Device Fault Diagnosis Method and Device

Abstract

The present disclosure relates to the technical field of fault diagnosis, and provides a device fault diagnosis method and apparatus. The method includes: obtaining device fault record data, device fault diagnosis data, and an expert diagnosis knowledge base; processing the device fault record data by using a device mechanism diagnosis model to obtain first diagnosis data; processing the device fault record data by using a knowledge graph diagnosis model to obtain second diagnosis data; processing the device fault record data by using a first neural network model to obtain third diagnosis data; annotating the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain an annotation result, and training a second neural network model by using the annotation result; constructing a diagnosis model by using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; and diagnosing a faulty device by using the diagnosis model.

Description

Technical Field

[0001] The present disclosure relates to the technical field of fault diagnosis, and in particular, to a method and device for diagnosing equipment faults. Background Art

[0002] Currently, for the method of diagnosing equipment faults, it is common to use the expert knowledge in the field of equipment faults to judge the fault causes of faulty equipment, such as fault trees, or use knowledge graphs to diagnose the fault causes of faulty equipment. However, whether using expert knowledge in the field of equipment faults or knowledge graphs to diagnose the fault causes of faulty equipment, there are problems of low efficiency and low accuracy.

[0003] In the process of implementing the concept of the present disclosure, the inventors found that there are at least the following technical problems in the related art: there are problems of low efficiency and low accuracy in diagnosing equipment faults. Summary of the Invention

[0004] In view of this, embodiments of the present disclosure provide a method and device for diagnosing equipment faults, an electronic device, and a computer-readable storage medium, so as to solve the problems of low efficiency and low accuracy in diagnosing equipment faults in the prior art.

[0005] In a first aspect of the embodiments of the present disclosure, a method for diagnosing equipment faults is provided, including: obtaining equipment fault record data, equipment fault diagnosis data, and an expert diagnosis knowledge base; constructing an equipment mechanism diagnosis model through an equipment mechanism analysis method based on the expert diagnosis knowledge base, and using the equipment mechanism diagnosis model to process the equipment fault record data to obtain first diagnosis data; constructing a knowledge graph diagnosis model through a knowledge graph reasoning method based on the equipment fault diagnosis data, and using the knowledge graph diagnosis model to process the equipment fault record data to obtain second diagnosis data; processing the equipment fault record data through a first neural network model to obtain third diagnosis data, where the first neural network model has been trained and learned and saved the corresponding relationship between the equipment fault record data and the third diagnosis data; annotating the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain an annotation result, and using the annotation result to train a second neural network model; constructing a diagnosis model using the second neural network model, the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; diagnosing a faulty equipment using the diagnosis model.

[0006] In a second aspect of the embodiments of the present disclosure, a device fault diagnosis apparatus is provided, including: an acquisition module configured to acquire device fault record data, device fault diagnosis data, and an expert diagnosis knowledge base; a first processing module configured to construct a device mechanism diagnosis model by means of a device mechanism analysis method based on the expert diagnosis knowledge base, and process the device fault record data by using the device mechanism diagnosis model to obtain first diagnosis data; a second processing module configured to construct a knowledge graph diagnosis model by means of a knowledge graph reasoning method based on the device fault diagnosis data, and process the device fault record data by using the knowledge graph diagnosis model to obtain second diagnosis data; a third processing module configured to process the device fault record data by using a first neural network model to obtain third diagnosis data, wherein the first neural network model has been trained to learn and store the correspondence between the device fault record data and the third diagnosis data; a training module configured to label the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain a labeling result, and use the labeling result to train a second neural network model; a construction module configured to construct a diagnosis model by using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; and a diagnosis module configured to diagnose a faulty device by using the diagnosis model.

[0007] In a third aspect of the embodiments of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the above method are implemented.

[0008] In a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0009] The beneficial effects of the embodiments of the present disclosure compared with the prior art are as follows: obtaining device failure record data, device failure diagnosis data, and an expert diagnosis knowledge base; based on the expert diagnosis knowledge base, constructing a device mechanism diagnosis model through a device mechanism analysis method, and using the device mechanism diagnosis model to process the device failure record data to obtain first diagnosis data; based on the device failure diagnosis data, constructing a knowledge graph diagnosis model through a knowledge graph reasoning method, and using the knowledge graph diagnosis model to process the device failure record data to obtain second diagnosis data; processing the device failure record data through a first neural network model to obtain third diagnosis data, wherein the first neural network model has been trained to learn and store the corresponding relationship between the device failure record data and the third diagnosis data; annotating the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain an annotation result, and using the annotation result to train a second neural network model; constructing a diagnosis model using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; and diagnosing a faulty device using the diagnosis model. By adopting the above technical means, the problems of low efficiency and low accuracy in diagnosing device failures in the prior art can be solved, thereby improving the efficiency and accuracy of diagnosing device failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the accompanying drawings required for use in the embodiments or the description of the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other accompanying drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 is a schematic diagram of the application scenario of the embodiments of the present disclosure;

[0012] Figure 2 is a schematic flowchart of a device failure diagnosis method provided by the embodiments of the present disclosure;

[0013] Figure 3 is a schematic structural diagram of a device failure diagnosis device provided by the embodiments of the present disclosure;

[0014] Figure 4 is a schematic structural diagram of an electronic device provided by the embodiments of the present disclosure. DETAILED DESCRIPTION

[0015] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art should understand that the present disclosure can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from obscuring the description of the present disclosure.

[0016] A method and device for diagnosing equipment faults according to an embodiment of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0017] Figure 1 It is a schematic diagram of the application scenario of the embodiment of the present disclosure. The application scenario may include terminal devices 1, 2, and 3, a server 4, and a network 5.

[0018] The terminal devices 1, 2, and 3 can be hardware or software. When the terminal devices 1, 2, and 3 are hardware, they can be various electronic devices with a display screen and supporting communication with the server 4, including but not limited to smartphones, tablets, laptop computers, and desktop computers, etc.; when the terminal devices 1, 2, and 3 are software, they can be installed in the above-mentioned electronic devices. The terminal devices 1, 2, and 3 can be implemented as multiple software or software modules, or can also be implemented as a single software or software module, and the embodiments of the present disclosure do not limit this. Further, various applications can be installed on the terminal devices 1, 2, and 3, such as data processing applications, instant messaging tools, social platform software, search applications, shopping applications, etc.

[0019] The server 4 can be a server that provides various services. For example, it is a background server that receives requests sent by terminal devices with which it establishes a communication connection. The background server can receive and analyze requests sent by terminal devices and generate processing results. The server 4 can be a single server, or can also be a server cluster composed of several servers, or can also be a cloud computing service center, and the embodiments of the present disclosure do not limit this.

[0020] It should be noted that the server 4 can be hardware or software. When the server 4 is hardware, it can be various electronic devices that provide various services for the terminal devices 1, 2, and 3. When the server 4 is software, it can be multiple software or software modules that provide various services for the terminal devices 1, 2, and 3, or can also be a single software or software module that provides various services for the terminal devices 1, 2, and 3, and the embodiments of the present disclosure do not limit this.

[0021] The network 5 can be a wired network connected by coaxial cables, twisted pair wires, and optical fibers, or a wireless network that enables the interconnection of various communication devices without the need for wiring. For example, Bluetooth, Near Field Communication (NFC), Infrared, etc. The embodiments of the present disclosure do not limit this.

[0022] Users can establish a communication connection between the terminal devices 1, 2, and 3 and the server 4 via the network 5 to receive or send information, etc. It should be noted that the specific types, quantities, and combinations of the terminal devices 1, 2, and 3, the server 4, and the network 5 can be adjusted according to the actual requirements of the application scenario. The embodiments of the present disclosure do not limit this.

[0023] Figure 2 It is a schematic flowchart of a device fault diagnosis method provided by the embodiments of the present disclosure. Figure 2 The device fault diagnosis method can be executed by Figure 1 the terminal device or the server. As Figure 2 shown, the device fault diagnosis method includes:

[0024] S201, obtaining device fault record data, device fault diagnosis data, and an expert diagnosis knowledge base;

[0025] S202, based on the expert diagnosis knowledge base, constructing a device mechanism diagnosis model through a device mechanism analysis method, and using the device mechanism diagnosis model to process the device fault record data to obtain first diagnosis data;

[0026] S203, based on the device fault diagnosis data, constructing a knowledge graph diagnosis model through a knowledge graph reasoning method, and using the knowledge graph diagnosis model to process the device fault record data to obtain second diagnosis data;

[0027] S204, processing the device fault record data through a first neural network model to obtain third diagnosis data, where the first neural network model has been trained to learn and store the corresponding relationship between the device fault record data and the third diagnosis data;

[0028] S205, annotating the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain an annotation result, and using the annotation result to train a second neural network model;

[0029] S206, constructing a diagnosis model using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model;

[0030] S207, diagnosing the faulty device using the diagnosis model.

[0031] Device failure record data records a large amount of information when devices fail, including: status information of each device, parameter information, and information on failure events, etc. Device failure diagnosis data is data on multiple devices that have failed and for which the specific failure causes have been diagnosed. The scale of device failure diagnosis data can be smaller than that of device failure record data. Taking the device failure diagnosis data as "ontology knowledge", a large number of practical device diagnosis-related knowledge can be quickly generated through knowledge graph reasoning methods, and then a knowledge graph diagnosis model can be constructed based on the generated knowledge. Ontology contains the basic entities within a certain discipline and the relationships between entities, and is a general concept model for describing domain knowledge. Ontology knowledge includes: information on device failures in device failure diagnosis data and the corresponding failure causes of the device failures in device failure diagnosis data. The generated knowledge also includes: information on device failures and the corresponding failure causes of the device failures. The scale of the generated knowledge is much larger than that of device failure diagnosis data. The first neural network model and the second neural network model can be any common type of neural network model.

[0032] Annotating the first diagnosis data, the second diagnosis data, and the third diagnosis data can be to annotate whether each piece of diagnosis data in the first diagnosis data, the second diagnosis data, and the third diagnosis data is correct. Since the second neural network model is trained with the annotation results, the second neural network model can determine which type of device, which failure event, and which scenario, etc. are more accurately diagnosed by the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model.

[0033] According to the technical solution provided by the embodiments of the present disclosure, obtain device failure record data, device failure diagnosis data, and an expert diagnosis knowledge base; based on the expert diagnosis knowledge base, construct a device mechanism diagnosis model through a device mechanism analysis method, and use the device mechanism diagnosis model to process the device failure record data to obtain the first diagnosis data; based on the device failure diagnosis data, construct a knowledge graph diagnosis model through a knowledge graph reasoning method, and use the knowledge graph diagnosis model to process the device failure record data to obtain the second diagnosis data; process the device failure record data through the first neural network model to obtain the third diagnosis data, where the first neural network model has been trained and has learned and saved the correspondence between the device failure record data and the third diagnosis data; annotate the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain an annotation result, and use the annotation result to train the second neural network model; use the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model to construct a diagnosis model; use the diagnosis model to diagnose the faulty device. By adopting the above technical means, the problems of low efficiency and low accuracy in diagnosing device failures in the prior art can be solved, thereby improving the efficiency and accuracy of diagnosing device failures.

[0034] In step S202, based on the expert diagnosis knowledge base, an equipment mechanism diagnosis model is constructed through the equipment mechanism analysis method, including: based on the expert diagnosis knowledge base, determining multiple fault events and one or more cause events corresponding to each fault event; based on the multiple fault events and one or more cause events corresponding to each fault event, generating a fault tree through the equipment mechanism analysis method; and constructing an equipment mechanism diagnosis model according to the fault tree.

[0035] The expert diagnosis knowledge base includes a large amount of diagnostic knowledge in the field of equipment fault diagnosis, such as the possible fault causes corresponding to a fault event of a certain equipment. A fault tree is a special inverted tree-shaped logical causal relationship diagram, which describes the causal relationship between various events in the system using event symbols, logic gate symbols, and transfer symbols. Fault events include descriptions of the phenomena of equipment failures, and cause events include descriptions of the causes of equipment failures. Because there are redundant information, duplicate information, and conflicting information in the expert diagnosis knowledge base, the equipment mechanism analysis method can assist in the correspondence between fault events and cause events, making the correspondence between fault events and cause events more clear. Or the equipment mechanism analysis method can also be understood as the application of the expert diagnosis knowledge base. There are multiple correspondence relationships between fault events and cause events in the fault tree, so the equipment mechanism diagnosis model is based on the expert diagnosis knowledge base and uses mathematical formulas to realize the correspondence between multiple fault events and cause events. For example, when a fault event is input into the equipment mechanism diagnosis model, a corresponding cause event can be output.

[0036] In step S202, the equipment mechanism diagnosis model is used to process the equipment fault record data to obtain the first diagnosis data, including: determining the parameter data of each equipment according to the equipment fault record data, where the parameter data includes: low-pressure pressure value, heating low-pressure unloading value, opening value of the main expansion valve, and low-pressure pressure change value; and determining the diagnosis result of each equipment according to the parameter data of each equipment to obtain the first diagnosis data.

[0037] For example: When the low-pressure switch alarms when the unit of a device enters the defrost mode for defrosting, the parameter data of the device collected within 30 minutes before and after the alarm time point is obtained. If the low-pressure pressure during defrosting > the heating low-pressure unloading value and the opening value of the main expansion valve > 400 steps, the possible cause of the alarm may be an error in the IoT link; if the low-pressure pressure < the heating low-pressure unloading value and the opening value of the main expansion valve > 400 steps, the possible cause of the alarm may be damage or loose connection of the low-pressure switch, and the solution is to replace or check the loose connection part of the wire. If there is a large change value in the low-pressure pressure, it may be an error in the IoT link. If there is no large change value in the low-pressure pressure, that is, the low-pressure pressure value changes smoothly, it may be that the system cannot recognize the operation. An error in the IoT link can be an error in the IoT link.

[0038] In step S203, based on the device fault diagnosis data, a knowledge graph diagnosis model is constructed through a knowledge graph reasoning method, including: performing knowledge extraction processing on the device fault diagnosis data to obtain an extraction result; performing knowledge fusion processing on the extraction result to obtain a fusion result; performing knowledge processing on the extraction result to obtain a device knowledge graph; and constructing a knowledge graph diagnosis model based on the device knowledge graph.

[0039] Knowledge extraction processing includes: entity extraction, relationship extraction, and attribute extraction. Entity extraction: Also known as Named Entity Recognition (NER), it refers to automatically identifying named entities from text corpora. Relationship extraction: After entity extraction from the text corpus, a series of discrete named entities are obtained. To obtain semantic information, it is necessary to extract the association relationships between entities from the relevant corpus, and connect the entities through relationships to form a networked knowledge structure. Attribute extraction: Collecting attribute information of specific entities from different information sources.

[0040] The information obtained through knowledge extraction has the following two problems: the relationships between information are flattened, lacking hierarchy and logic; there is a large amount of redundant information. Knowledge fusion is used to solve the above problems and mainly includes two parts: entity linking and knowledge merging. Among them, entity linking involves two technologies: coreference resolution and entity disambiguation.

[0041] Knowledge processing mainly includes three aspects: ontology construction, knowledge reasoning, and quality assessment. Ontology contains the basic entities within a certain discipline and the relationships between entities, and is a general concept model for describing domain knowledge. After the ontology construction of the knowledge graph is completed, it has taken initial shape, but the relationships between knowledge are incomplete. Knowledge reasoning is used for further knowledge discovery to complete the knowledge of the knowledge graph. Knowledge reasoning is mainly divided into three categories: rule-based reasoning, graph-based reasoning, and deep learning-based reasoning. Quality assessment is also an important part of the knowledge base construction technology, used to quantify the credibility of knowledge, and to ensure the quality of the knowledge graph by discarding knowledge with low confidence.

[0042] The knowledge graph diagnosis model is to realize the corresponding relationship between multiple fault events and cause events based on the knowledge graph with the help of mathematical formulas. For example, when a fault event is input into the knowledge graph diagnosis model, a corresponding cause event can be output.

[0043] In step S203, the knowledge graph diagnosis model is used to process the device fault record data to obtain second diagnosis data, including: determining the device attribute information of each device and one or more fault events corresponding to each device according to the device fault record data; determining the diagnosis result of each device according to the device attribute information of each device and one or more fault events corresponding to each device, so as to obtain the second diagnosis data.

[0044] The device attribute information includes: the subordinate relationship of the device and the device group relationship. Because in a device cluster, the fault of one device is often related to other devices related to this device. In the embodiments of the present disclosure, with the help of the knowledge graph diagnosis model, the relationship between the device attribute information of each device, one or more fault events corresponding to each device, and the fault cause of each device is found.

[0045] Training the first neural network model can be deep learning training. Since model training is a prior art, it will not be elaborated here.

[0046] In step S206, a diagnosis model is constructed by using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model, including: forming a branch network by connecting the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model in parallel; using the second neural network model in front of the branch network to construct the diagnosis model.

[0047] In the diagnosis model, first use the second neural network model to judge which model among the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model has a high accuracy rate in diagnosing the device to be diagnosed according to the device information of the device to be diagnosed, and then use the judged model to diagnose the device to be diagnosed. The device information of the device to be diagnosed includes the type of the device, the historical fault information of the device, and the application scenario to which the device belongs.

[0048] In step S207, diagnosing the faulty device by using the diagnosis model, including: obtaining the device information of the faulty device; processing the device information by using the second neural network model to obtain a processing result, where the processing result is used to determine a target model, and the target model includes: the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; processing the device information by using the target model to obtain the diagnosis result of the faulty device.

[0049] Optionally, the processing result can also be used to represent the weights among the diagnosis results of the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model in equipment fault diagnosis. Then, the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model are respectively used to process the equipment information to obtain three diagnosis results, and then based on the three diagnosis results and the weights corresponding to the three diagnosis results respectively, the final diagnosis result is confirmed.

[0050] All the above optional technical solutions can be combined arbitrarily to form optional embodiments of the present application, which will not be elaborated here one by one.

[0051] The following is an embodiment of the apparatus of the present disclosure, which can be used to execute the embodiment of the method of the present disclosure. For details not disclosed in the embodiment of the apparatus of the present disclosure, please refer to the embodiment of the method of the present disclosure.

[0052] Figure 3 It is a schematic diagram of an equipment fault diagnosis apparatus provided by an embodiment of the present disclosure. As Figure 3 shown, the equipment fault diagnosis apparatus includes:

[0053] An acquisition module 301, configured to acquire equipment fault record data, equipment fault diagnosis data, and an expert diagnosis knowledge base;

[0054] A first processing module 302, configured to construct an equipment mechanism diagnosis model based on the expert diagnosis knowledge base through an equipment mechanism analysis method, and use the equipment mechanism diagnosis model to process the equipment fault record data to obtain first diagnosis data;

[0055] A second processing module 303, configured to construct a knowledge graph diagnosis model based on the equipment fault diagnosis data through a knowledge graph reasoning method, and use the knowledge graph diagnosis model to process the equipment fault record data to obtain second diagnosis data;

[0056] A third processing module 304, configured to process the equipment fault record data through a first neural network model to obtain third diagnosis data, where the first neural network model has been trained to learn and store the corresponding relationship between the equipment fault record data and the third diagnosis data;

[0057] A training module 305, configured to label the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain a labeling result, and use the labeling result to train a second neural network model;

[0058] A construction module 306, configured to construct a diagnosis model by using the second neural network model, the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model;

[0059] A diagnosis module 307, configured to diagnose a faulty equipment by using the diagnosis model.

[0060] The device failure record data records a large amount of information when the device fails, including: the status information of each device, parameter information, and information on failure events, etc. The device failure diagnosis data is data on multiple devices that have failed and for which the specific failure causes have been diagnosed. The scale of the device failure diagnosis data can be smaller than that of the device failure record data. Taking the device failure diagnosis data as "ontology knowledge", a large number of practical device diagnosis-related knowledge can be quickly generated through knowledge graph reasoning methods, and then a knowledge graph diagnosis model can be constructed based on the generated knowledge. Ontology contains the basic entities within a certain discipline and the relationships between entities, and is a general concept model for describing domain knowledge. Ontology knowledge includes: the information on device failures in the device failure diagnosis data and the corresponding failure causes of the device failures in the device failure diagnosis data. The generated knowledge also includes: the information on device failures and the corresponding failure causes of the device failures. The scale of the generated knowledge is much larger than that of the device failure diagnosis data. The first neural network model and the second neural network model can be any common neural network model.

[0061] Annotating the first diagnostic data, the second diagnostic data, and the third diagnostic data can be to annotate whether each piece of diagnostic data in the first diagnostic data, the second diagnostic data, and the third diagnostic data is correct. Since the second neural network model is trained with the annotation results, the second neural network model can determine which type of device, which failure event, and which scenario, etc. are more accurately diagnosed by the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model.

[0062] According to the technical solution provided by the embodiments of the present disclosure, device fault record data, device fault diagnosis data, and an expert diagnosis knowledge base are obtained; based on the expert diagnosis knowledge base, a device mechanism diagnosis model is constructed by a device mechanism analysis method, and the device mechanism diagnosis model is used to process the device fault record data to obtain first diagnosis data; based on the device fault diagnosis data, a knowledge graph diagnosis model is constructed by a knowledge graph reasoning method, and the knowledge graph diagnosis model is used to process the device fault record data to obtain second diagnosis data; the device fault record data is processed by a first neural network model to obtain third diagnosis data, where the first neural network model has been trained to learn and store the corresponding relationship between the device fault record data and the third diagnosis data; the first diagnosis data, the second diagnosis data, and the third diagnosis data are labeled to obtain a labeling result, and the labeling result is used to train a second neural network model; a diagnosis model is constructed by using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; the diagnosis model is used to diagnose a faulty device. By adopting the above technical means, the problems of low efficiency and low accuracy in diagnosing device faults in the prior art can be solved, thereby improving the efficiency and accuracy of diagnosing device faults.

[0063] Optionally, the first processing module 302 is further configured to determine a plurality of fault events and one or more cause events corresponding to each fault event based on the expert diagnosis knowledge base; generate a fault tree by a device mechanism analysis method based on the plurality of fault events and one or more cause events corresponding to each fault event; and construct a device mechanism diagnosis model according to the fault tree.

[0064] The expert diagnosis knowledge base includes a large amount of diagnosis knowledge in the field of device fault diagnosis, such as the possible fault causes corresponding to a fault event of a device. A fault tree is a special inverted tree-shaped logical causal relationship diagram, which uses event symbols, logic gate symbols, and transfer symbols to describe the causal relationship between various events in the system. The fault event includes a description of the phenomenon of the device failure, and the cause event includes a description of the cause of the device failure. Because there is redundant information, duplicate information, and conflicting information in the expert diagnosis knowledge base, the device mechanism analysis method can assist in the correspondence between the fault event and the cause event, making the correspondence between the fault event and the cause event clearer. Or the device mechanism analysis method can also be understood as the application of the expert diagnosis knowledge base. There are multiple correspondences between the fault event and the cause event in the fault tree, so the device mechanism diagnosis model is based on the expert diagnosis knowledge base and realizes the correspondence between the multiple fault events and the cause events with the help of mathematical formulas. For example, when a fault event is input to the device mechanism diagnosis model, a corresponding cause event can be output.

[0065] Optionally, the first processing module 302 is further configured to determine parameter data of each device according to device fault record data, where the parameter data includes: low-pressure pressure value, heating low-pressure unloading value, opening value of the main expansion valve, and low-pressure pressure change value; and determine a diagnosis result of each device according to the parameter data of each device to obtain first diagnosis data.

[0066] For example: when a low-pressure switch alarm occurs during defrosting of a device unit, obtain the parameter data of the device collected within 30 minutes before and after the alarm time point. If the low-pressure pressure during defrosting > heating low-pressure unloading value and the opening value of the main expansion valve > 400 steps, the alarm reason may be an error in the IoT link; if the low-pressure pressure < heating low-pressure unloading value and the opening value of the main expansion valve > 400 steps, the alarm reason may be damage or loose connection of the low-pressure switch, and the solution is to replace or check the loose connection part of the wire. If there is a large change in the low-pressure pressure value, it may be an error in the IoT link; if there is no large change in the low-pressure pressure value, that is, the low-pressure pressure value changes smoothly, it may be that the system cannot recognize the operation. An error in the IoT link can be an error in the IoT link.

[0067] Optionally, the second processing module 303 is further configured to perform knowledge extraction processing on the device fault diagnosis data to obtain an extraction result; perform knowledge fusion processing on the extraction result to obtain a fusion result; perform knowledge processing on the extraction result to obtain a device knowledge graph; and construct a knowledge graph diagnosis model based on the device knowledge graph.

[0068] Knowledge extraction processing includes: entity extraction, relationship extraction, and attribute extraction. Entity extraction: Also known as Named Entity Recognition (NER), it refers to automatically identifying named entities from text corpora. Relationship extraction: After entity extraction from the text corpus, a series of discrete named entities are obtained. To obtain semantic information, it is necessary to extract the association relationships between entities from the relevant corpus and connect the entities through relationships to form a networked knowledge structure. Attribute extraction: Collect attribute information of specific entities from different information sources.

[0069] The information obtained through knowledge extraction has the following two problems: the relationships between information are flattened, lacking hierarchy and logic; there is a large amount of redundant information. Knowledge fusion is used to solve the above problems, mainly including two parts: entity linking and knowledge merging, where entity linking involves two techniques: coreference resolution and entity disambiguation.

[0070] Knowledge processing mainly includes three aspects: ontology construction, knowledge reasoning, and quality assessment. Ontology contains the basic entities within a certain discipline and the relationships between entities, and is a general conceptual model for describing domain knowledge. After the ontology construction of the knowledge graph is completed, it has taken initial shape, but the relationships between knowledge are incomplete. Knowledge reasoning is used to further discover knowledge, so as to complete the knowledge in the knowledge graph. Knowledge reasoning is mainly divided into three categories: rule-based reasoning, graph-based reasoning, and deep learning-based reasoning. Quality assessment is also an important part of the knowledge base construction technology, which is used to quantify the credibility of knowledge and ensure the quality of the knowledge graph by discarding knowledge with low confidence.

[0071] The knowledge graph diagnosis model is to realize the corresponding relationship between multiple fault events and cause events based on the knowledge graph with the help of mathematical formulas. For example, if a fault event is input into the knowledge graph diagnosis model, a corresponding cause event can be output.

[0072] Optionally, the second processing module 303 is further configured to determine the device attribute information of each device and one or more fault events corresponding to each device according to the device fault record data; determine the diagnosis result of each device according to the device attribute information of each device and one or more fault events corresponding to each device, so as to obtain the second diagnosis data.

[0073] The device attribute information includes: the subordinate relationship of the device and the device group relationship. Because in a device cluster, the fault of one device is often related to other devices related to this device. In the embodiments of the present disclosure, with the help of the knowledge graph diagnosis model, the relationship between the device attribute information of each device, one or more fault events corresponding to each device, and the fault cause of each device is found.

[0074] Training the first neural network model can be deep learning training. Since model training is a prior art, it will not be elaborated here.

[0075] Optionally, the construction module 306 is further configured to form a branch network by connecting the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model in parallel; use the branch network to connect to the second neural network model in front to construct the diagnosis model.

[0076] In the diagnosis model, first use the second neural network model to judge which model among the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model has a high accuracy in diagnosing the device to be diagnosed according to the device information of the device to be diagnosed, and then use the judged model to diagnose the device to be diagnosed. The device information of the device to be diagnosed includes the type of the device, the historical fault information of the device, and the application scenario to which the device belongs.

[0077] Optionally, the diagnosis module 307 is further configured to obtain device information of a faulty device; process the device information by using a second neural network model to obtain a processing result, where the processing result is used to determine a target model, and the target model includes: a device mechanism diagnosis model, a knowledge graph diagnosis model, and a first neural network model; and process the device information by using the target model to obtain a diagnosis result of the faulty device.

[0078] Optionally, the processing result can also be used to represent weights among the diagnosis results of the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model in device fault diagnosis. Then, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model are respectively used to process the device information to obtain three diagnosis results, and then based on the three diagnosis results and the weights respectively corresponding to the three diagnosis results, the final diagnosis result is confirmed.

[0079] It should be understood that the sequence numbers of the steps in the above embodiments do not indicate the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present disclosure.

[0080] Figure 4 is a schematic diagram of the electronic device 4 provided by the embodiment of the present disclosure. As Figure 4 shown, the electronic device 4 of this embodiment includes: a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. When the processor 401 executes the computer program 403, the steps in the above various method embodiments are implemented. Alternatively, when the processor 401 executes the computer program 403, the functions of each module / unit in the above various device embodiments are implemented.

[0081] Exemplarily, the computer program 403 can be divided into one or more modules / units, and one or more modules / units are stored in the memory 402 and executed by the processor 401 to complete the present disclosure. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 403 in the electronic device 4.

[0082] The electronic device 4 can be a desktop computer, a notebook, a palm computer, a cloud server, and other electronic devices. The electronic device 4 can include, but is not limited to, the processor 401 and the memory 402. Those skilled in the art can understand that Figure 4 merely examples of the electronic device 4 do not constitute a limitation to the electronic device 4, and it can include more or fewer components than shown in the figure, or combine some components, or different components. For example, the electronic device can further include input / output devices, network access devices, a bus, etc.

[0083] The processor 401 can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0084] The memory 402 can be an internal storage unit of the electronic device 4, for example, the hard disk or memory of the electronic device 4. The memory 402 can also be an external storage device of the electronic device 4, for example, a plug-in hard disk equipped on the electronic device 4, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 402 can also include both the internal storage unit and the external storage device of the electronic device 4. The memory 402 is used to store computer programs and other programs and data required by the electronic device. The memory 402 can also be used to temporarily store data that has been output or will be output.

[0085] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0086] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0087] Those of ordinary skill in the art will realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this disclosure.

[0088] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus / electronic device and method can be implemented in other ways. For example, the apparatus / electronic device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. Multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the apparatus or unit can be in electrical, mechanical or other forms.

[0089] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, the functional units in each embodiment of this disclosure can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0091] When the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present disclosure, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. The computer program can include computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0092] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A method for diagnosing equipment faults, characterized in that, it includes: Obtain equipment fault record data, equipment fault diagnosis data, and an expert diagnosis knowledge base; Based on the expert diagnosis knowledge base, construct an equipment mechanism diagnosis model through the equipment mechanism analysis method, and use the equipment mechanism diagnosis model to process the equipment fault record data to obtain first diagnosis data; Based on the equipment fault diagnosis data, construct a knowledge graph diagnosis model through the knowledge graph reasoning method, and use the knowledge graph diagnosis model to process the equipment fault record data to obtain second diagnosis data; Process the equipment fault record data through a first neural network model to obtain third diagnosis data, where the first neural network model has been trained to learn and store the correspondence between the equipment fault record data and the third diagnosis data; Annotate the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain an annotation result, and use the annotation result to train a second neural network model; Use the second neural network model, the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model to construct a diagnosis model; Use the diagnosis model to diagnose the faulty equipment; Among them, the constructing an equipment mechanism diagnosis model through the equipment mechanism analysis method based on the expert diagnosis knowledge base includes: based on the expert diagnosis knowledge base, determine multiple fault events and one or more cause events corresponding to each fault event; based on the multiple fault events and one or more cause events corresponding to each fault event, generate a fault tree through the equipment mechanism analysis method; construct an equipment mechanism diagnosis model according to the fault tree; Among them, the constructing a knowledge graph diagnosis model through the knowledge graph reasoning method based on the equipment fault diagnosis data includes: perform knowledge extraction processing on the equipment fault diagnosis data to obtain an extraction result; perform knowledge fusion processing on the extraction result to obtain a fusion result; perform knowledge processing on the extraction result to obtain an equipment knowledge graph; construct a knowledge graph diagnosis model based on the equipment knowledge graph; Among them, the constructing a diagnosis model using the second neural network model, the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model includes: forming a branch network by connecting the equipment mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model in parallel; using the branch network to connect the second neural network model in front to construct a diagnosis model.

2. The method according to claim 1, characterized in that, the using the equipment mechanism diagnosis model to process the equipment fault record data to obtain first diagnosis data includes: According to the equipment fault record data, determine the parameter data of each device, where the parameter data includes: low-pressure pressure value, heating low-pressure unloading value, opening value of the main expansion valve, and low-pressure pressure change value; According to the parameter data of each device, determine the diagnosis result of each device to obtain first diagnosis data.

3. The method according to claim 1, characterized in that, Processing the device fault record data by using the knowledge graph diagnosis model to obtain second diagnosis data, including: Determining device attribute information of each device and one or more fault events corresponding to each device according to the device fault record data; Determining a diagnosis result of each device according to the device attribute information of each device and one or more fault events corresponding to each device to obtain second diagnosis data.

4. The method according to claim 1, wherein, The diagnosing the faulty device by using the diagnosis model includes: Obtaining device information of the faulty device; Processing the device information by using the second neural network model to obtain a processing result, wherein the processing result is used to determine a target model, and the target model includes: the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; Processing the device information by using the target model to obtain a diagnosis result of the faulty device.

5. A device fault diagnosis device, wherein, including: An acquisition module configured to acquire device fault record data, device fault diagnosis data, and an expert diagnosis knowledge base; A first processing module configured to construct a device mechanism diagnosis model by using a device mechanism analysis method based on the expert diagnosis knowledge base, and process the device fault record data by using the device mechanism diagnosis model to obtain first diagnosis data; A second processing module configured to construct a knowledge graph diagnosis model by using a knowledge graph reasoning method based on the device fault diagnosis data, and process the device fault record data by using the knowledge graph diagnosis model to obtain second diagnosis data; A third processing module configured to process the device fault record data by using a first neural network model to obtain third diagnosis data, wherein the first neural network model has been trained to learn and store the corresponding relationship between the device fault record data and the third diagnosis data; A training module configured to label the first diagnosis data, the second diagnosis data, and the third diagnosis data to obtain a labeling result, and train a second neural network model by using the labeling result; A construction module configured to construct a diagnosis model by using the second neural network model, the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model; A diagnosis module configured to diagnose a faulty device by using the diagnosis model; The first processing module is further configured to determine a plurality of fault events and one or more cause events corresponding to each fault event based on the expert diagnosis knowledge base; generate a fault tree by using a device mechanism analysis method based on the plurality of fault events and one or more cause events corresponding to each fault event; construct a device mechanism diagnosis model according to the fault tree; The second processing module is further configured to perform knowledge extraction processing on the device fault diagnosis data to obtain an extraction result; perform knowledge fusion processing on the extraction result to obtain a fusion result; perform knowledge processing on the extraction result to obtain a device knowledge graph; and construct a knowledge graph diagnosis model based on the device knowledge graph. The construction module is further configured to form a branch network by connecting the device mechanism diagnosis model, the knowledge graph diagnosis model, and the first neural network model in parallel; and connect the second neural network model in front of the branch network to construct a diagnosis model.

6. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, when the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium storing a computer program, wherein, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images